21:00:23 #startmeeting quantum 21:00:24 Meeting started Mon Mar 11 21:00:23 2013 UTC. The chair is danwent. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:25 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:26 hi 21:00:28 The meeting name has been set to 'quantum' 21:00:37 #info agenda: https://wiki.openstack.org/wiki/Network/Meetings 21:00:53 #topic announcements 21:01:09 first off, congrats to markmcclain, who will be the new Quantum PTL :) 21:01:22 congrats! 21:01:30 markmcclain: i have a bunch of emails I haven't replied to yet for quantum... i'm thinking of just forwarding them to you :P 21:01:50 #info markmcclain will be quantum ptl for havana 21:01:50 markmcclain: congratulations sr. 21:02:03 all hail our new lider maximo! 21:02:23 #info the grizzly-rc1 branch date is tomorrow, march 12th, EOD pacific time 21:02:39 hi 21:02:50 once we pull that branch, master will be open for havana commits, and we will remove -2 from existing reviews 21:03:09 danwent: yong said that he will not be able to make the meeting tonight 21:03:12 Just a heads up, I will be OOO wed-sun 21:03:16 garyk: thx 21:03:50 #info at team meeting next monday, we will decide if we need to release an RC2, and if so, what changes we would need to backport from master 21:04:11 so if you have a fix post RC1, we will first commit it to master and backport (much like an existing stable release) 21:04:23 any other announcements/comments? 21:04:44 #topic Grizzly RC1 bugs (markmcclain) 21:05:03 #info https://launchpad.net/quantum/+milestone/grizzly-rc1 21:05:17 First, thanks to you dan.. you've left some big shoes to fill as PTL 21:05:46 markmcclain: thanks. time to take me behind the barn and shoot me :P 21:06:09 Is THAT the transition process. 21:06:19 med_: no body told you? 21:06:31 ok markmcclain , looks like we have 5 active bugs? 21:06:39 haha 21:06:52 yeah and 3 of those are all related to migrations 21:06:59 https://bugs.launchpad.net/quantum/+bug/1153585 21:07:00 Launchpad bug 1153585 in quantum "Add a migration for plugins which added L3 support in Grizzly" [High,In progress] 21:07:07 https://bugs.launchpad.net/quantum/+bug/1153366 21:07:09 Launchpad bug 1153366 in quantum "port_security migration does not migrate data" [High,In progress] 21:07:14 https://bugs.launchpad.net/quantum/+bug/1153616 21:07:16 Launchpad bug 1153616 in quantum "nvp_port_mapping breaks upgrade from folsom for NVP users" [High,In progress] 21:07:48 they're currently covered by two core devs each and we're testing the fixes 21:08:25 the two medium fixes have reviewers but I'm not sure they've been tested enough 21:08:26 https://bugs.launchpad.net/quantum/+bug/1115999 21:08:28 Launchpad bug 1115999 in quantum "quantum-netns-cleanup does not stop metadata proxies" [Medium,In progress] 21:09:31 amotoki or garyk any thoughts on this? 21:09:58 review is here: https://review.openstack.org/#/c/21489/ 21:10:05 markmcclain: it looked good last time i saw it - minor little change 21:10:31 ok… so that will stay in for RC1 21:10:42 markmcclain: i think that tomorrow it will be done 21:10:53 https://bugs.launchpad.net/quantum/+bug/1135948 21:10:55 Launchpad bug 1135948 in quantum "Wrong behavior when a DHCP agent goes down" [Medium,In progress] 21:11:16 yeah, both of those patchsets seemed a bit on the big side, but as long as we have core devs testing + reviewing, I think they are ok 21:11:44 fixes a problem that existed in Folsom too… the change has been simplified a great deal 21:12:08 salv-orlando: do you still have time available to review it? 21:12:13 yep 21:12:22 ok.. so we've got 2 cores on it 21:12:33 that's it for the bugs…anything else? 21:12:56 markmcclain: is this a bug that should be backported? 21:12:56 markmcclain: we have been seeing issues in quantum/nova integration when using an admin toke 21:13:00 token 21:13:12 but those should be changes purely to nova, I suspect. 21:13:26 ok.. is there a bug to track? 21:13:31 have another bug to mention in python quantumclient section 21:13:34 danwent: can you please elaborate 21:14:29 garyk: need to do the full bug write-up, but basically, nova/quantum integration currently seems to require that you use a "service user", rather than an "admin token" 21:15:21 it seems like we should be doing something more akin to how nova authenticates for talking to glance 21:15:28 danwent: ok. i know that certain requests require to admin t get port details. the rest should be regulra user 21:15:49 which uses a single set of auth config in nova, rather than quantum-specific auth variables. 21:16:31 garyk: anyway, I don't know as much about this as I should yet... will file a bug when i get done working through it myself. 21:16:41 ok, thanks 21:16:57 garyk: could be that i'm just missing something... i tend to just configure nova to use a service user 21:17:10 ok, anything else on RC1 bugs. 21:17:19 danwent: yeah, me too 21:17:51 #topic quantum python-client 21:18:14 we released 2.2.0 on Mar 7th 21:18:15 markmcclain: again? 21:18:51 arosen: has proposed requirement updates for the other projects to raise the minimum to 2.2.0 21:18:57 markmcclain: markmcclain nevermind, was just saying you need to talk again :) 21:19:14 I get tired of hearing me talk too :) 21:19:31 the requirement update process has a gone a little slower than expected 21:20:15 we're the first requirement change that's going through the centralized requirements process so everyone is figuring out how the process works 21:21:26 the client does have one bug that was reported shortly before the meeting that we'll look at 21:21:27 https://bugs.launchpad.net/python-quantumclient/+bug/1153715 21:21:29 Launchpad bug 1153715 in python-quantumclient "no commandline option to set insecure=True" [High,New] 21:21:30 markmcclain: this came up today: https://bugs.launchpad.net/python-quantumclient/+bug/1153715 21:21:37 haha 21:21:53 my shitty IRC webclient doesn't scroll... so I keep missing that you've already said things. 21:22:49 danwent: 21:22:50 web chat on free node usually scrolls 21:22:51 seems like openstack projects need to do a better job of making sure "generic" flags like this as applied across all projects 21:23:10 salv-orlando: mmm... that's what i'm using 21:24:00 any other client questions? 21:24:32 nope 21:24:49 #topic documentation (danwent) 21:24:57 so it seems like we're a bit behind on docs. 21:25:13 i took a full pass through the admin guide and have started updating general items 21:25:37 I particularly need people to help around metadata and lbaas 21:26:23 salv-orlando: what is status on API guide updates? I think you'll need to wrestle more work out of people than I will for the admin guide. 21:26:44 danwent: I'll claim the metadata doc ticket 21:26:45 I am starting doing by myself the provider net updates 21:26:59 which is simple. But I really need help for the LBaaS chapter which is huge 21:27:11 danwent: i need to document the vif updates 21:27:29 I'm also going to randomly select one between nati-ueno and arosen for the security group API work! 21:27:30 any takers for the lbaas? otherwise, I'll just do a varient of the wiki page I wrote to make sure there is something there 21:27:41 salv-orlando: Thanks! 21:28:12 congrats nati_ueno who has just accepted to the the API doc for security groups! 21:28:18 (you did not read the fine print) 21:28:38 salv-orlando: He he I'll do my best 21:28:40 Ok, at the meeting next week, we need all doc bugs to be filed. We've done a great job getting the code in shape, now we need to make sure the docs are also up to par. 21:29:05 markmcclain: i also need help on the folsom + db upgrade stuff 21:29:26 right… I'll help with that too 21:29:28 I want to make sure people are testing this, as I think its a likely source of bugs, and I don't want those bugs popping up too late in the cycle 21:29:48 (we already saw a flurry of migration related bugs in the last few days). 21:30:19 an advert... there is pagination api doc ready for review at https://review.openstack.org/#/c/22536/ 21:30:23 I can take a crack at the initial metadata stuff if that frees you up to do the DB 21:31:00 ok, anything else on docs? 21:31:00 also an advert… review https://review.openstack.org/#/c/20960/ so that we can publish to api.openstack.org/api-ref.html 21:31:21 salv-orlando: has been reviewing, maybe it's good-to-go, just need that last + 21:31:31 salv-orlando: do you have already bug report for security group stuff? 21:31:37 salv-orlando: thanks for reviewing that already 21:31:52 nati_ueno: I need to find it gimme a sec 21:32:08 ok, moving on to horizon 21:32:12 danwent: actually annegentle had to chase me a lot as I kept forgetting that review :/ 21:32:24 #topic Horizon + quantum ( amotoki / nati_ueno ) 21:32:27 yeah it's hard to remember the api-site repo :) it's okay 21:32:45 congrat for Akihiro. He is also horizon core now! 21:32:57 congratulations 21:33:04 +2 21:33:12 nati_ueno: thanks. It help us to bridge QUantum and Horizon. 21:33:19 everyone thanks! 21:33:33 congrats amotoki! 21:33:38 nice! 21:33:56 congrats amotoki 21:33:57 amtoki, congratulations 21:34:05 also, if you guys haven't seen the new net visualization stuff from nati_ueno in grizzly horizon, it is VERY cool 21:34:20 indeed! great work nachi! 21:34:21 makes giving quantum demos much more fun. 21:34:26 danwent: Thanks! also from Hayashi from my team 21:34:43 that's cool 21:35:01 anther topic is lbaas. quantum-lbaas has been merged and minor fixed/improvements are under reivew. 21:35:32 we received a new bug https://bugs.launchpad.net/horizon/+bug/1153477 yesterday. 21:35:36 Launchpad bug 1153477 in horizon "editing security group rules ends up with ValueError when Quantum secgroup is enabled" [Undecided,New] 21:36:15 I need to wait to post a fix until next week. 21:36:30 since I will take off for a week from this Wed. 21:37:00 amotoki: ok, or maybe someone else can take a crack at it as well. 21:37:20 I hope so, but currently not.... 21:37:28 i will talk to arosen, as he has been making the nova security group proxy changes 21:37:40 #topic open discussion 21:37:43 It's truly horizon issue. 21:37:51 any other topics to discuss? 21:37:58 amotoki: yup 21:38:36 please send email to core with any new issues that may impact RC1 21:38:59 again, I will be mostly (completely according to my wife) offline from wed :) 21:39:34 danwent: enjoy 21:39:38 ok, nice short meeting :) 21:39:43 I'm pretty sure she will succeed 21:39:58 ok, thanks again for all of the reviewing in the last few days... really put us in good shape 21:39:58 +1 for danwent's wife :) 21:40:11 nati_ueno: haha 21:40:23 ok, talk to you all later. have a good afternoon/evening :) 21:40:26 #endmeeting